Skip to main content
. 2012 Jan 12;28(5):656–663. doi: 10.1093/bioinformatics/bts028

Table 4.

Comparison of the different methods on tissue-specific enhancers

Performance on tissue-specific enhancer sequences
5% Precision
AUC ROC
AUC PR
Tissue F M L H F M L H F M L H
D2 0.61 0.64 0.55 0.50 0.55 0.55 0.50 0.45 0.54 0.55 0.51 0.47
D2z 0.66 0.69 0.63 0.56 0.57 0.57 0.56 0.53 0.57 0.57 0.55 0.52
D2* 0.71 0.70 0.67 0.60 0.62 0.62 0.59 0.55 0.60 0.60 0.58 0.54
N2* 0.65 0.64 0.62 0.58 0.58 0.57 0.56 0.53 0.57 0.56 0.55 0.53
N2rc 0.71 0.67 0.68 0.60 0.61 0.59 0.58 0.55 0.60 0.58 0.58 0.55
N2mm(1.0),rc 0.84 0.82 0.79 0.66 0.66 0.64 0.63 0.57 0.66 0.64 0.63 0.57

Bold numbers indicate the best performance. Positive sequences were obtained by ChIP-Seq of p300 in forebrain (F), midbrain (M), limb (L) and heart (H) tissue of the mouse embryo. Negative sequences were randomly sampled from the mouse genome. All pairwise scores were computed with repeats masked, k=6, background Markov model of order 1. Results show average values over 25 samples each time drawing 500 sequences.

HHS Vulnerability Disclosure